Comments (8)
by threshold I mean the "-t" argument, this specifies the confidence level on whether the model considers a voxel point to be a part of the binding site or not. If the model does not predict any points with confidence greater than "t" you will not get any segmented shapes for that pocket.
from deeppocket.
yes thats right
from deeppocket.
I have looked at several examples, and they all look like this. That cube has only one or two dots. I did not edit the code, and this is what I got.
1a9w_E.clean_nowat_1.txt
1a52_A.clean_nowat_1.txt
1aii_A.clean_nowat_1.txt
from deeppocket.
the .dx files are supposed to be predicted outputs from the segmentation model put in that format for visualization. Sometimes the model doesn't like the pocket provided to it and so it predicts very little, if this is a persisting problem then it is worrisome. Just for debugging purposes, could you use a lower threshold and see if you get a larger set of points?
from deeppocket.
in any case, even if the segmentation model doesn't like the pocket, the classification model still ranks the fpocket output. So you could use the reranked fpocket predictions if that serves your purpose
from deeppocket.
@RishalAggarwal, by lower threshold, do you mean a higher distance? >3.5? Is there an email address where I can reach you? I have another query I would like to discuss! Thanks!
from deeppocket.
The .dx issue is actually for ALL pockets. Those segmented out, and those that are not.
from deeppocket.
I see, but then the confidence of the segmented pockets would be lower, right?
from deeppocket.
Related Issues (20)
- Training Classifier Problem HOT 1
- Transform part in Train HOT 5
- lack of gninatypes files HOT 9
- The number of pocket in types file HOT 3
- .
- predict.py running error HOT 2
- Understanding the Train Dataset for Training Part HOT 2
- Pocket Probability HOT 1
- Channels in training script are different from those in your Supporting Information table s1 HOT 6
- Do not have files for running make_types.py when prerparing custom data for training a new classifier HOT 3
- Training Classifier Dataset HOT 8
- can not find seg0_best_test_IOU_91.pth.tar.
- No output after running train_segmentation.py HOT 6
- Broken Link to Model Checkpoints, "404 FILE NOT FOUND" HOT 2
- Issue with archived file HOT 3
- molgrid version? HOT 2
- Yaml file or Colab Notebook? HOT 2
- Why does DeepPocket not predict residues for some fpocket pockets?
- Predicting a Binding Site (predict.py): TypeError: sequence item 0: expected str instance, numpy.float32 found HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from deeppocket.